Although chemotherapy has significantly improved the survival of breast cancer patients, treatment failure still remains a major clinical issue worldwide. Current knowledge about treatment failure is mostly derived from research on intrinsic and acquired chemoresistance in epithelial tumor cells. However, recent studies have implicated a critical role for host cells (i.e., the tissue microenvironment) in building a protective ?niche? for tumor cells enabling their escape from chemotherapeutic treatments. Notably, the host regenerative response upon chemotherapy ?injury?, which is regarded as a host intrinsic mechanism to repair damaged tissues, may be exploited by tumor cells for their local recurrence or distant metastases. The proposed study will build on our extensive experience in the study of tumor-stroma interactions in breast cancer metastasis, and will investigate the poorly explored question of how chemotherapy-induced changes in the lung stroma foster the early relapse of tumor cells in the lung. Based on our previous findings that tissue resident mesenchymal stem cells (MSCs) acquire a significantly higher potential to promote local tumor growth upon cancer therapies, we hypothesize that systemic chemotherapy treatment stimulates regenerative responses in lung resident MSCs, which are, in turn, utilized by drug-resistant disseminated tumor cells (DTCs) for their metastatic relapse in the lung. By designing different chemotherapy treatment scenarios in animal models mimicking clinical situations in human breast cancer patients, we will in Aim 1 determine how chemotherapeutic drugs cisplatin and doxorubicin modulate the lung resident MSCs using our newly established endogenous MSC modeling platform in mice. Subsequently, we will in Aim 2 delineate the molecular mechanisms underlying drug-activated lung resident MSCs to support metastatic tumor growth in the lung, with a focus on the TLR4 signaling pathway and the key wound healing cytokine osteopontin (OPN). Finally, we will in Aim 3 define the translational potential of stroma targeting approaches using both patient-derived xenograft models and breast cancer patient specimen analyses. We will specifically focus on the therapeutic efficacy of combining chemotherapy with TLR4 or OPN blockage in treatment of patient-derived human basal-like xenograft breast tumors. Further, by analyzing clinical plasma samples from human breast cancer patients, we expect to develop plasma OPN as a biomarker to predict early metastatic relapse of breast cancer patients after chemotherapy. Overall, our proposed study will energize an underdeveloped field of research that investigates the impact of cancer therapeutics on the pre-metastatic microenvironment. Findings from the proposed study will facilitate the development of clinically applicable strategies to improve treatment efficacy and prevent metastatic relapse of breast cancer by interfering with the tissue metastatic microenvironment.